Selective Direct Mode Encoding for Video Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing direct mode encoding methods for moving images suffer from reduced coding efficiency due to inaccurate motion vector derivation, especially when temporal correlation is greater than spatial correlation, and when the same position block of the backward reference picture is encoded as an intra block, leading to increased coding bits and decreased prediction capability.
Innovation Solution
A selective direct mode encoding method that calculates motion vectors using both temporal and spatial correlations, selects the optimal mode based on image correlation characteristics, and derives bidirectional motion vectors using equations tailored for each mode, ensuring efficient motion vector coding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If temporal direct mode encoding is used to derive motion vectors from backward reference pictures, then motion vector compression efficiency is improved, but coding efficiency deteriorates when temporal correlation is greater than spatial correlation or when the same position block is encoded as intra block
Solution Approach 1:
The patent implements dynamic mode selection that adapts between temporal direct mode, spatial direct mode, and skip mode based on picture type, block position, and correlation characteristics. The encoder dynamically determines the optimal direct mode for each block rather than using a fixed temporal direct mode, thereby resolving the contradiction between compression efficiency and derivation accuracy under varying conditions
Solution Approach 2:
The patent changes the parameter of motion vector derivation by introducing multiple direct mode variants (temporal, spatial, and their combinations) and selectively applying them based on correlation analysis. This parameter change allows the system to switch between different derivation strategies to maintain accuracy while preserving compression benefits
2Manufacturing precision
If spatial direct mode encoding is used to derive motion vectors from neighboring blocks, then coding efficiency is improved under certain conditions, but overall coding efficiency decreases when temporal correlation dominates
Solution Approach 1:
The patent implements dynamic mode selection that adapts between temporal direct mode, spatial direct mode, and skip mode based on picture type, block position, and correlation characteristics. The encoder dynamically determines the optimal direct mode for each block rather than using a fixed spatial direct mode, thereby resolving the contradiction between derivation accuracy and coding efficiency under varying conditions
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
Disclosed is a method of encoding a direct mode when coding moving images, the method includes finding a prediction and a cost for each of a plurality of direct modes including a backward-based temporal direct mode and a forward-based temporal direct mode with respect to a current block to be encoded and determining one mode having the lowest cost as a result of comparing the costs as an optimal encoding mode, thereby performing encoding the direct mode.